Driving risk intelligent early warning method, system and equipment considering vehicle interaction

Abstract

The invention discloses a driving risk intelligent early warning method, system and device considering vehicle interaction. The method comprises the following steps: establishing a driving risk assessment model; historical vehicle trajectory data are collected, key features are extracted for clustering, clustering results are divided into risk levels, and early warning thresholds corresponding to the risk levels are determined; real-time vehicle motion state information of the vehicle and the front vehicle is obtained in real time, so that a real-time risk level is obtained; and issuing corresponding preset early warning information to the driver according to the real-time risk level. The driving risk assessment model is utilized to accurately assess the potential collision risk of the vehicle in real time, a large amount of historical vehicle trajectory data is combined, the early warning threshold values of different risk levels are determined, effective early warning suggestions are provided in different risk scenes, and a driver is better assisted to cope with the potential risk. According to the method, the accuracy and the real-time performance of risk assessment can be improved, the driving safety in a complex traffic environment is remarkably improved, and the accident risk is reduced.

Description

Technical Field

[0001] The present invention relates to the field of traffic warning technology, and in particular, to a method, system and device for intelligent warning of driving risks taking vehicle interaction into consideration. Background Art

[0002] With the popularization of modern transportation and the increasing complexity of road traffic, the complex and changeable traffic flow environment has put forward higher requirements for driving safety. Traditional driving safety warning systems often rely on the driver's experience and simple judgment of vehicle sensors, and it is difficult to comprehensively and accurately assess potential collision risks. In order to improve driving safety, a more scientific and accurate risk assessment method is urgently needed. The driving safety warning methods in the prior art mainly include the following methods:

[0003] 1. Driving safety warning methods that directly identify risks based on sensors, such as detecting obstacles ahead and road coverings through driving safety warning devices, and calculating recommended speeds based on the distance to the obstacles; or obtaining location information of surrounding vehicles through image data, and generating warning information when the distance between vehicles is less than a preset distance. This type of method is simple to implement and has high warning efficiency, but it is difficult to use for warnings in complex traffic scenes and emergencies. At the same time, sensors are difficult to perceive potential collision risks and have a certain lag.

[0004] 2. Driving safety warning methods based on risk indicators and prediction models, such as predicting the severity of accidents between public buses and surrounding vehicles and the relative probability of accidents to obtain driving risk values, and assessing risks in combination with the driver's risk feedback; or assessing the risk of human-vehicle-road coupling by constructing the probability of collision accidents between pedestrians and vehicles and the probability of skidding and rollover accidents. If the risk assessment result exceeds the threshold, an early warning is issued. This type of method can accurately assess risks and issue early warnings in a timely manner, but the model of this type of method is relatively complex and requires high computational costs.

[0005] In summary, the driving safety warning methods in the existing technology are mainly based on direct risk identification by sensors or risk assessment and prediction model assessment and driving safety warning. However, these methods are difficult to accurately identify risks in a complex and changeable road traffic environment with a simple and fast method. Summary of the invention

[0006] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes an intelligent warning method for driving risk considering vehicle interaction, which can accurately simulate the dynamic relationship between vehicles and achieve more accurate and effective driving safety warning in complex traffic environments.

[0007] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0008] An intelligent warning method for driving risk considering vehicle interaction comprises the following steps: S1, establishing a driving risk assessment model; S2, collecting historical vehicle trajectory data, extracting key features representing risks in the historical trajectory data for clustering, dividing the clustering results into different risk levels, and determining the corresponding warning thresholds for each risk level according to the average risk value of samples of each risk level; S3, acquiring real-time vehicle motion state information of the vehicle and the preceding vehicle in real time, and calculating the real-time vehicle risk value according to the driving risk assessment model, and judging the risk level corresponding to the real-time vehicle risk value according to the warning thresholds of each risk level to obtain the real-time risk level; S4, issuing corresponding preset warning information to the driver according to the real-time risk level.

[0009] Furthermore, the driving risk assessment model f(s) is as follows:

[0010]

[0011] where s e is the critical safety distance between the vehicle and the vehicle in front; s is the actual distance between the vehicle and the vehicle in front; D is the equilibrium bond length s e The potential energy is minimum at ; α is the vibration frequency response coefficient.

[0012] Furthermore, where s e The calculation formula is as follows:

[0013] In the formula, SSD p is the safe stopping distance of the vehicle ahead, SSD f is the safe stopping distance of the vehicle; h is the center distance between the front vehicle and the vehicle; v p and v f are the speeds of the preceding vehicle and the vehicle itself; L p and L f are the lengths of the preceding vehicle and the vehicle itself respectively; a p and a f are the maximum decelerations of the preceding vehicle and the vehicle itself respectively; PRT is the perception reaction time.

[0014] Furthermore, in step S2, the key features characterizing the risk include: the speed of the vehicle, the speed of the preceding vehicle, the distance and speed difference between the vehicle and the preceding vehicle, and the risk value calculated based on a driving risk assessment model.

[0015] Furthermore, in step S2, the risk is first divided into k levels, and then the warning thresholds of each risk level are determined respectively; the warning thresholds corresponding to each risk level are determined according to the average risk value of samples of each risk level, specifically by the following formula: Among them, R kis the warning threshold of the k-th level risk; n k is the number of samples of the k-th risk level; r i is the risk value of the i-th sample in the k-th risk level.

[0016] Furthermore, in step S2, a clustering algorithm is used to cluster the historical trajectory data, wherein the number of clusters is k.

[0017] Furthermore, in step S3, the real-time vehicle motion state information includes the vehicle type of the vehicle, static information of the vehicle length of the vehicle, the speed of the vehicle, the vehicle type of the preceding vehicle, static information of the vehicle length of the preceding vehicle, the speed of the preceding vehicle, and the vehicle spacing.

[0018] Furthermore, the warning information includes display information of the vehicle display screen and voice prompt information of the voice broadcast system; the risk level with a lower warning threshold has a higher prompt frequency and intensity.

[0019] The present invention also provides a driving collision avoidance risk intelligent transportation Internet of Things warning system that takes vehicle interaction into consideration, including: a risk assessment model establishment module, used to establish a driving risk assessment model; a clustering module, used to collect historical vehicle trajectory data, extract key features characterizing risks in the historical trajectory data for clustering, and divide the clustering results into different risk levels; a threshold determination module, used to determine the corresponding warning thresholds for each risk level according to the average risk value of samples of each risk level; a data acquisition module, used to obtain real-time vehicle motion state information of the vehicle and the preceding vehicle; a risk assessment module, used to calculate the real-time vehicle risk value according to the real-time vehicle motion state information, and judge the risk level corresponding to the real-time vehicle risk value according to the warning thresholds of each risk level to obtain the real-time risk level; and a warning issuance module, used to issue corresponding preset warning information to the driver according to the real-time risk level.

[0020] The present invention also proposes an electronic device, including a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to execute the intelligent warning method for driving risks taking into account vehicle interactions.

[0021] The present invention has the following beneficial effects:

[0022] The driving risk assessment model is used to accurately and in real time assess the potential collision risk of vehicles. Combined with a large amount of historical vehicle trajectory data, the warning thresholds of different risk levels are determined, and effective warning suggestions are provided in different risk scenarios to better assist drivers in dealing with potential risks. The present invention can improve the accuracy and real-time performance of risk assessment, significantly improve driving safety in complex traffic environments, and reduce accident risks.

[0023] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0025] Figure 1 It is a schematic diagram of the overall process of the method of the present invention;

[0026] Figure 2 is a schematic diagram of the system of the present invention;

[0027] Figure 3 It is a schematic diagram of a driving risk assessment model based on molecular dynamics provided in an embodiment of the present invention.

[0028] Figure 4 This is a risk clustering result diagram of historical trajectory data provided by an embodiment of the present invention.

[0029] Figure 5 It is a sample vehicle risk value curve diagram provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0030] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0033] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0034] Please refer to Figure 1 In a preferred embodiment of the present invention, an intelligent warning method for driving risk taking vehicle interaction into consideration includes steps S1, S2, S3 and S4.

[0035] S1, establish a driving risk assessment model; refer to Figure 3 , which is a schematic diagram of a driving risk assessment model based on molecular dynamics. Since the molecular interactions in microscopic particle systems have systematic similarities with the interaction behaviors between vehicles in the macroscopic world, vehicles are regarded as molecules, and the molecular interaction forces are used to describe the interaction relationship between vehicles. The driving risk assessment model is obtained by improving the molecular interaction force function. By introducing the principles of molecular dynamics, the dynamic relationship between vehicles can be accurately simulated, which can achieve more accurate and effective driving safety warnings in complex traffic environments.

[0036] S2, collect historical vehicle trajectory data, extract key features representing risks from historical trajectory data for clustering, divide the clustering results into different risk levels, and determine the corresponding warning thresholds for each risk level based on the average risk value of samples of each risk level. Figure 4 , which is a risk clustering result diagram of historical trajectory data provided by an embodiment of the present invention.

[0037] S3, real-time acquisition of the real-time vehicle motion status information of the vehicle and the preceding vehicle, and calculation of the real-time vehicle risk value based on the driving risk assessment model, and judge the risk level corresponding to the real-time vehicle risk value based on the warning thresholds of various risk levels to obtain the real-time risk level; specifically, the real-time vehicle motion status information of the vehicle and the preceding vehicle can be collected in real time through GPS, radar, camera and other sensing devices, and the data is transmitted to the central processor for data preprocessing by the processor, the driving risk assessment model and threshold are preset in the processor, the processor calculates the vehicle risk value to obtain the real-time vehicle risk value of the vehicle, and obtains the real-time risk level based on the threshold. Reference Figure 5 , which is a sample vehicle risk value curve diagram of an embodiment of the present invention, obtains the real-time vehicle risk value and obtains the real-time vehicle risk value according to the change of the driving time Figure 5 The graph shown.

[0038] S4, issuing corresponding preset warning information to the driver according to the real-time risk level. The warning information is generated by the processor according to the real-time risk level. The warning information needs to be preset. Specifically, the warning information can be issued to the driver in a timely manner through a display screen or voice prompt.

[0039] The driving risk assessment model is obtained by improving the molecular interaction force function. The driving risk assessment model is expressed as f(s). The function value is the driving risk value. When f(s)<0, it indicates that the actual distance between the vehicle and the vehicle in front is less than the critical safety distance, and there is a collision risk. Specifically, the driving risk assessment model f(s) is as follows:

[0040]

[0041] where s e is the critical safety distance between the vehicle and the vehicle in front; s is the actual distance between the vehicle and the vehicle in front; D is the equilibrium bond length s e The minimum potential energy at , i.e., the energy required for bond breaking, reflects the potential risk of the vehicle and the preceding vehicle when maintaining the critical safety distance, which can be understood as the potential risk coefficient of the vehicle and the preceding vehicle when maintaining the critical safety distance; α is the vibration frequency response coefficient, which is a parameter related to the vibration frequency; e is a natural constant, i.e., the base of the natural logarithm. D and α are constants. For example, in this embodiment, D is 3.2 and α is 0.38.

[0042] In a specific embodiment of the present invention, wherein s e The calculation formula is as follows:

[0043]

[0044]

[0045]

[0046] In the formula, SSD p is the safe stopping distance of the vehicle ahead, SSD f is the safe stopping distance of the vehicle; h is the center distance between the front vehicle and the vehicle; v p and v f are the speeds of the preceding vehicle and the vehicle itself; L p and L f are the lengths of the preceding vehicle and the vehicle itself respectively; a p and a f are the maximum decelerations of the preceding vehicle and the vehicle itself, respectively, and are determined according to the maximum decelerations specified for different vehicle types in relevant standards; PRT is the perception reaction time, and in a specific embodiment of the present invention, the PRT may be 1.5s.

[0047] In some embodiments of the present invention, in step S2, the key features characterizing the risk include: the speed of the vehicle, the speed of the preceding vehicle, the distance and speed difference between the vehicle and the preceding vehicle, and the risk value calculated based on the driving risk assessment model. In a specific embodiment of the present invention, the extracted key features are exemplified in the following table:

[0048]

[0049] In some embodiments of the present invention, in step S2, the warning threshold corresponding to each risk level is determined according to the average risk value of samples of each risk level, specifically by the following formula:

[0050]

[0051] Among them, R k is the warning threshold of the k-th level risk; n k is the number of samples of the k-th risk level; r i is the risk value of the i-th sample in the k-th level risk. Specifically, the extracted feature values ​​are standardized, and the number of clusters is set to 3, k = 1, 2, 3. In step S2, a clustering algorithm is used to cluster the historical trajectory data, and the obtained clustering results are divided into low risk, medium risk and high risk, k = 1, 2, 3 represent low risk, medium risk and high risk respectively, and the clustering algorithm can be a K-Means clustering algorithm or other clustering algorithms.

[0052] In a specific embodiment of the present invention, the early warning thresholds of various levels of risk obtained by clustering are as follows:

[0053] Risk Level Warning threshold Low risk 0.51 Medium risk -0.10 High risk -7.72

[0054] When the real-time vehicle risk value is between 0.51 and -0.10, it is assessed as low risk; when the real-time vehicle risk value is between -0.10 and -7.72, it is assessed as medium risk; when the real-time vehicle risk value is less than -7.72, it is assessed as high risk. When the real-time vehicle risk value is equal to the warning threshold, it is assessed as the risk level corresponding to the warning threshold. For example, when the real-time vehicle risk value is 0.51, it is assessed as low risk, and so on.

[0055] In a specific embodiment of the present invention, in step S3, the real-time vehicle motion state information includes the vehicle type of the vehicle, static information of the vehicle length of the vehicle, the speed of the vehicle, the vehicle type of the preceding vehicle, static information of the vehicle length of the preceding vehicle, the speed of the preceding vehicle, and the vehicle spacing. The length L of the vehicle can be obtained according to the vehicle type of the vehicle. f and the maximum deceleration a f , according to the type of the preceding vehicle, the length L of the preceding vehicle can be obtained pand the maximum deceleration a p , and then according to the static information of the vehicle length, the speed of the vehicle and the static information of the vehicle length and the speed of the vehicle in front, the real-time vehicle risk value can be obtained according to the driving risk assessment model. In addition, each sensor stores the acquired real-time vehicle motion state information to the processor, which integrates the original data and performs data preprocessing such as removing noise and outliers. Data preprocessing specifically includes missing value completion, duplicate data removal, noise processing, and outlier detection and processing.

[0056] Missing value completion: If there are missing values, for vehicle type and vehicle length, the incomplete record can be filled in based on the previous adjacent record or deleted; for the vehicle speed, the speed of the preceding vehicle, and the vehicle spacing, the missing records are completed using the mean of the two adjacent points before and after.

[0057] De-duplicate data: Delete duplicate records and retain unique records to ensure data independence.

[0058] Noise processing: For velocity data, the Savitzky-Golay filtering method is used to smooth noise and reduce data fluctuations. This method is based on the local polynomial least squares fitting, where the fitting polynomial p(j) is used to fit the data points in the sliding window. The least squares method minimizes the residual sum of squares ε in the sliding window. N To determine the coefficients a of the polynomial j , specifically through the following formula:

[0059]

[0060]

[0061] Among them, a n is the polynomial coefficient; N is the order of the polynomial, n is the term of the polynomial, and represents the power of j in the polynomial; j is the position index in the sliding window, j n represents the nth power of position index j; ε N is the residual sum of squares; x[j] is the original data point at position j in the window; M is half the width of the sliding window, and the window size is 2M+1; p(j) is the fitting value of the fitted polynomial at n; in some embodiments, N=3, M=3.

[0062] Outlier detection and processing: The standard deviation method is used to detect outliers. When |x-μ|>3σ, that is, the difference between the data x and the mean μ exceeds 3 standard deviations σ, it is considered an outlier, and the mean of the two adjacent points before and after the missing value is used to fill in the missing value.

[0063] In a specific embodiment of the present invention, the warning information includes display information of the vehicle display screen and voice prompt information of the voice broadcast system; the risk level with a lower warning threshold has a higher prompt frequency and intensity, that is, as the risk level increases, the prompt frequency and intensity are increased.

[0064] In a specific embodiment of the invention, the warning is shown in the following table:

[0065]

[0066] Reference Figure 2 The present invention also proposes an intelligent transportation Internet of Things warning system for traffic collision avoidance risk taking into account vehicle interaction, including a risk assessment model building module, a clustering module, a threshold determination module, a data acquisition module, and a risk assessment module.

[0067] The risk assessment model building module is used to build a driving risk assessment model.

[0068] The clustering module is used to collect historical vehicle trajectory data, extract key features representing risks in the historical trajectory data for clustering, and divide the clustering results into different risk levels.

[0069] The threshold determination module is used to determine the corresponding warning thresholds for each risk level according to the average risk value of samples of each risk level, and transmit the warning thresholds for each level of risk to the risk assessment module.

[0070] The data acquisition module is used to obtain the real-time vehicle motion status information of the vehicle and the preceding vehicle. The GPS collects the position and speed of the vehicle in real time. The on-board radar collects the speed and relative position of the preceding vehicle in real time. The camera identifies the vehicle type and length of the preceding vehicle.

[0071] The risk assessment module is used to calculate the real-time vehicle risk value based on the real-time vehicle motion status information, and determine the risk level corresponding to the real-time vehicle risk value based on the warning thresholds of various risk levels to obtain the real-time risk level, and transmit the real-time risk level to the warning release module.

[0072] The warning release module is used to release corresponding preset warning information to the driver according to the real-time risk level.

[0073] In order to improve the quality of data, an intelligent transportation Internet of Things warning system for traffic collision avoidance risk taking into account vehicle interaction also includes a data processing module. The information collected by the data acquisition module is transmitted to the data processing module. The data processing module integrates and stores the received raw data and performs data cleaning to provide high-quality data input for subsequent risk assessment.

[0074] The present invention also provides an electronic device, including a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to execute the intelligent warning method for driving risks taking into account vehicle interactions.

[0075] It should be understood that in the embodiment of the present invention, the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the type of device.

[0076] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent early warning method for driving risks considering vehicle interaction, It is characterized in that The steps include: S1, establish a driving risk assessment model; S2, collect historical vehicle trajectory data, extract key features representing risks from the historical trajectory data for clustering, divide the clustering results into different risk levels, and determine the corresponding warning thresholds for each risk level based on the average risk value of samples of each risk level; S3, real-time acquisition of vehicle motion status information of the vehicle and the preceding vehicle, and calculation of the real-time vehicle risk value according to the driving risk assessment model, and determination of the risk level corresponding to the real-time vehicle risk value according to the early warning thresholds of various risk levels, so as to obtain the real-time risk level; S4, issuing corresponding preset warning information to the driver according to the real-time risk level.

2. The intelligent early warning method for driving risk considering vehicle interaction according to claim 1, It is characterized in that The driving risk assessment model f(s) is as follows: where s e is the critical safety distance between the vehicle and the vehicle in front; s is the actual distance between the vehicle and the vehicle in front; D is the equilibrium bond length s e The potential energy is minimum at ; α is the vibration frequency response coefficient.

3. The intelligent early warning method for driving risk considering vehicle interaction according to claim 2, It is characterized in that where s e The calculation formula is as follows: In the formula, SSD p is the safe stopping distance of the vehicle ahead, SSD f is the safe stopping distance of the vehicle; h is the center distance between the front vehicle and the vehicle; v p and v f are the speeds of the preceding vehicle and the vehicle itself; L p and L f are the lengths of the preceding vehicle and the vehicle itself respectively; a p and a f are the maximum decelerations of the preceding vehicle and the vehicle itself respectively; PRT is the perception reaction time.

4. The intelligent early warning method for driving risk considering vehicle interaction according to claim 1, It is characterized in that In step S2, the key features characterizing the risk include: the speed of the vehicle, the speed of the preceding vehicle, the distance and speed difference between the vehicle and the preceding vehicle, and the risk value calculated based on the driving risk assessment model.

5. The intelligent early warning method for driving risk considering vehicle interaction according to claim 3, It is characterized in that In step S2, the warning threshold corresponding to each risk level is determined according to the average risk value of samples of each risk level, which is specifically performed by the following formula: Among them, R k is the warning threshold of the k-th level risk; n k is the number of samples of the k-th risk level; r i is the risk value of the i-th sample in the k-th risk level.

6. The intelligent early warning method for driving risk considering vehicle interaction according to claim 4, It is characterized in that In step S2, a clustering algorithm is used to cluster the historical trajectory data.

7. The intelligent early warning method for driving risk considering vehicle interaction according to claim 5, It is characterized in that In step S3, the real-time vehicle motion state information includes the vehicle type of the vehicle, static information of the vehicle length of the vehicle, the speed of the vehicle, the vehicle type of the preceding vehicle, static information of the vehicle length of the preceding vehicle, the speed of the preceding vehicle, and the vehicle spacing.

8. The intelligent early warning method for driving risk considering vehicle interaction according to claim 1, It is characterized in that The warning information includes display information on the vehicle display screen and voice prompt information from the voice broadcast system; the risk level with a lower warning threshold has a higher corresponding prompt frequency and intensity.

9. An intelligent traffic IoT early warning system for collision avoidance risk considering vehicle interaction. It is characterized in that include: Risk assessment model building module, used to build a driving risk assessment model; The clustering module is used to collect historical vehicle trajectory data, extract key features representing risks from the historical trajectory data for clustering, and divide the clustering results into different risk levels; A threshold determination module is used to determine the corresponding warning thresholds for each risk level according to the average risk value of samples of each risk level; Data acquisition module, used to obtain real-time vehicle motion status information of the vehicle and the preceding vehicle; The risk assessment module is used to calculate the real-time vehicle risk value according to the real-time vehicle motion state information, and determine the risk level corresponding to the real-time vehicle risk value according to the warning thresholds of various risk levels to obtain the real-time risk level; The warning release module is used to release corresponding preset warning information to the driver according to the real-time risk level.

10. An electronic device comprising a processor and a memory, wherein the memory is used to store a computer program. It is characterized in that The processor is used to execute the computer program to execute the intelligent warning method for driving risk considering vehicle interaction as described in any one of claims 1 to 8.

Images